Forgot your password?
typodupeerror
The Almighty Buck Software Linux

GPL 3.0 to Penalize Google, Amazon? 582

Posted by Zonk
from the fine-print dept.
Michael Ferris writes "Is this the start of a shakedown by the GNU folks? Michael Singer writes that Eben Moglen and the folks rewriting the GPL are looking at a proposal where companies would be required to pay money if they use GPLed software, even if they don't redistribute the software." From the article: "The current version of the GPL, which was last updated in 1991, fails to trigger the open source license if a company alters the code, but does not distribute its software through a CD or floppy disk...the [current] rule does not apply to companies that distribute software as a service, such as Google and eBay, or even dual-license companies like Sleepycat."
This discussion has been archived. No new comments can be posted.

GPL 3.0 to Penalize Google, Amazon?

Comments Filter:
  • by caluml (551744) <slashdot@spCOWam ... minus herbivore> on Friday April 08, 2005 @05:58PM (#12181803) Homepage
    If I am ever to write something worthy of releasing to the world (and not just something I am playing around with), I will explicitly specify which version of the GPL I am releasing it under.
    Call me paranoid, but I wouldn't want even the faintest chance that some nasty corporation managed to litigate itself in the position of being able to release a future GPL version, as in bold below:

    This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version.

    That's handing control of the licencing of your code over to whoever is allowed to write GPLv1851, if I'm reading it correctly.
    • That's handing control of the licencing of your code over to whoever is allowed to write GPLv1851, if I'm reading it correctly.

      I'm worried about GPLv1984 myself.

    • by KiloByte (825081) on Friday April 08, 2005 @06:11PM (#12181964)
      RMS is a madman. Fortunately, he's our madman.
      He needs to be watched closely (to prevent blunders like the GFDL), but, he is well-known to have good intentions.

      The problem is, if something bad happens to him, it's possible that whatever members of FSF will have the deciding say will push the GPL in a completely different direction. Whoever controls the FSF, controls the vast majority of GPLed software.

      I'm not paranoid enough to label FSFians as possible traitors who would follow whoever shakes the purse, hell no -- I have quite a bit of faith in them. However, they may do any modifications to the licenses of software they don't own the copyright to -- it's a huge power. It's dangerous to leave such power in the hands of people not protected by insanity.
      • I'm not paranoid enough to label FSFians as possible traitors who would follow whoever shakes the purse, hell no

        Oh, I agree completely - I don't think they'd do something like that. However, they may be forced to hand over control to another party. It's unlikely, but stranger legal things have happened.

      • by Janek Kozicki (722688) on Friday April 08, 2005 @07:08PM (#12182472) Journal
        No, the best part of this comment is the closing sentence, not the opening one:

        It's dangerous to leave such power in the hands of people not protected by insanity.

        most clever coclusion I've heard in years. Funny also because of the inside pardox. But very true.
      • Stop your FUD (Score:3, Interesting)

        by JoeBuck (7947)
        The FSF board consists of respected people like Eben Moglen and Larry Lessig. They aren't going to allow the betrayal of the FSF.
      • Whoever controls the FSF, controls the vast majority of GPLed software.

        Why are you saying such horrible things?

        Do you believe that if v4 said "you must send $1,000 the FSF for every execution of the program," that your v2 or v3 code would suddenly be v4? That your users would suddenly be required to send money to the FSF?

        Why do you believe this?
    • by ecklesweb (713901) on Friday April 08, 2005 @06:27PM (#12182128)
      Well, I do have software distributed under the GPL, so I want to talk specifics...

      Here's the full section 9, a portion of which you quoted:

      9. The Free Software Foundation may publish revised and/or new versions of the General
      Public License from time to time. Such new versions will be similar in spirit to the
      present version, but may differ in detail to address new problems or concerns.

      Each version is given a distinguishing version number. If the Program specifies a
      version number of this License which applies to it and "any later version", you have
      the option of following the terms and conditions either of that version or of any
      later version published by the Free Software Foundation. If the Program does not
      specify a version number of this License, you may choose any version ever published by
      the Free Software Foundation.


      So my question becomes this: What determines whether or not you specified a specific version of the GPL? Most of my comments and the readme file say "Licensed under the GNU General Public License" (no version mentioned), but then included with the distribution is a copy of version 2 of the license. Does that imply strongly enough that version 2 is the specific license under which the software is distributed?

      Or do I need to go make some changes and do a commit....

      • by Dysan2k (126022) on Friday April 08, 2005 @06:45PM (#12182281) Homepage
        Well, if you license under GPL, you're supposed to include the LICENSE file which includes the GPL in it's entirity. Now, if you haven't been including the license in your package, that's a problem you need to remedy. The license itself has the version number at the top, so it's easy to identify. You may want to clarify from here on out, which version you are licensing under, but from the usual standpoint, it falls something like this:

        Developer: It doesn't have a version number, so it refers to the earliest version
        Business: It doesn't have a version number, so it must be the most recent
        Reality: It has no version number, so it would apply to the current version upon creation of the software
        Legal: Pay me lots o' cash, and I'll make something up and back it in court. *Crosses-fingers*
      • Anyone who writes software licensed under the GPL should read the section at the bottom entitled "How to Apply These Terms to Your New Programs". It tells you exactly what you need to do.
    • by ajs (35943) <ajs AT ajs DOT com> on Friday April 08, 2005 @06:32PM (#12182176) Homepage Journal
      A few points for sanity here:

      1. No one can force a company to abide by the rules of a new version of the GPL for software the currently have under an old version. They can CHOOSE to apply the new version IF the author used the standard boilerplate license notice, but they can also CHOOSE not to
      2. The article specifically states that there is no GPLv3 and they're not officially comenting on what they are considering for it when it does come out
      3. The guy commenting is saying what he would like to see so that he can drop his unique license. That's fine, but it's not official word
      4. If this were to happen in the doomsday sense, everyone woudl immediately fork old copies of the programs that they have the option to apply the GPLv2 to, and continue to maintain and license them as such.

    • > (at your option)

      I you can use older of newer version. This is a very smart move by RMS. If say there is a new situation in the world he'll be able to give people options with new licences, hopefully not allowing for regular style licences(commercial, no source).
      Get over it. :)
      2c.

      PS: drinking with Alan Cox is something entirely different
    • by Magic Thread (692357) on Friday April 08, 2005 @07:30PM (#12182663) Homepage Journal
      Copyright law doesn't cover use of the software - only distribution. If someone is not distributing the software, copyright law for the software does not apply. So how can you make a service provider release code for something they aren't distributing?

      For this to work as described, the user would have had to sign a contract. Otherwise, it will be just as unenforceable as a EULA (you already had a right to click the Next button on your own computer, etc).
  • They deliver HTML. (Score:5, Insightful)

    by The Amazing Fish Boy (863897) on Friday April 08, 2005 @05:58PM (#12181806) Homepage Journal
    the [current] rule does not apply to companies that distribute software as a service, such as Google and eBay

    Google and eBay distribute HTML. That HTML is created by software that uses GPL code. So if I modify a GPL Office Suite, would I have to distribute the code if I email someone a document I made with it? Seems like a bad idea, in general.

    I guess people could fork the GPL2.0'd code if the software developers switched to GPL 3.0
    • by ScentCone (795499) on Friday April 08, 2005 @06:00PM (#12181843)
      Google and eBay distribute HTML

      Hmm. I could swear that the Google appliance in my rack, and the Google toolbar on my desktop weren't just hunks of HTML.
      • Hmm. I could swear that the Google appliance in my rack, and the Google toolbar on my desktop weren't just hunks of HTML.

        Google Toolbar - Not GPL.

        Google appliance - By the terms of the GPL, they already have to distribute the code if they modify a GPL program. (I'm pretty sure.)
        • No they have to offer you a reasonable way to get the code if they *distrubute* the changes. So under the current version of the GPL if they are shipping GPLed code on the appliance in question they have to offer you a way to get the source.

          But you are basically right and the GP is clearly wrong.
        • by Richard_at_work (517087) <richardprice AT gmail DOT com> on Friday April 08, 2005 @06:18PM (#12182042)
          Has it every been proven that the google appliance is a Linux box, because Ive seen one in the flesh and played with it on a network, and it most certainly looks like a unix box of some description (nmap identified it as a FreeBSD 4 server among other things) as of 6 months ago.

          Just because they use Linux in the Googleplex doesnt mean they use it everywhere.
          • by NetNifty (796376)
            Well here, [theregister.co.uk] they say that "According to Google, the 1U Linux box can handle 60 queries a minute...", although a quick look at the google appliance site doesnt say its a Linux box, although the article is over 3 years old.
      • What does that have to do with anything? Do those things use GPL code?
    • by raehl (609729) <raehl311 AT yahoo DOT com> on Friday April 08, 2005 @06:58PM (#12182388) Homepage
      Google and eBay allow end users to run programs on Google and eBay servers that create HTML for the end user. If I go to Google and do a search, INPUT occurs on my computer, PROCESSING occurs on Google's computer, and OUTPUT occurs on my computer.

      That's not the same as you distributing a document you created. In that case, INPUT occurs on your computer, PROCESSING occurs on your computer, and OUTPUT occurs on your computer.

      The question is: What's important in determining if a program has been "distributed"? Clearly if I give you a CD with the program and you run it on your computer, I've distributed the program. Clearly if I take GPL software and modify it and I run it on my computer only, I have not distributed the program.

      But if I take GPL'd software, modify it, and then let YOU run it on my computer...

      Did I distribute it, or not?

      Your immediate answer is probably "Who cares?" But now what if I charge you to run this modified program on my computer, for example, by charging a fee if you use my auction program? Now I'm using software provided by the open source community for my financial gain, but not returning the modifications I made to that software to the community.

      The extreme of this problem is that eventually, the internet becomes so fast and clients become so dumb that software is never "distributed" at all. I take an open source office suite and then modify it. According to the GPL, if I then sell that software on CD, or by download, so that people can actually run it on their computers, I must provide the source to it as well. But what if I'd rather just make money off of the GPL'd software I've taken without giving anything back to the community?

      Well, then I just put the software on my own server, and instead of selling CDs or downloads, I let people provide input and receive output over a remote connection to the program running on my server.

      And wala! People can modify and essentially provide GPL'd software without having to provide source.
      • It's "voila"
  • by dtolton (162216) * on Friday April 08, 2005 @05:59PM (#12181815) Homepage
    I think requiring companies to pay is a big time mistake, and to me, it
    hails back to the days of the old BSD style licenses. The GPL works
    so well now, precisely because it is unobtrusive. My company runs
    GPL'd software because we are able to use it and make modification
    without either redistributing the source code and we aren't required
    to pay for that ability.

    Requiring people to open all their changes or pay for them will put a
    lot of businesses off when it comes to dealing with GPL'd software. I
    don't think that is a good trade off to make, and I don't think it
    will be healthy for the open source community in general.

    A move like this will make the newer BSD style licenses and / or
    licenses like the Python license much more attractive imo. Now that
    open source is finally turning the corner, and solid technologies are
    finally moving into the enterprise, why would we even entertain making
    changes that will certainly hamper open source adoption?

    This isn't a consistent position in my opinion. If you are
    developing free software, it should remain free. The idea that you
    can get someone hooked on software, and then pull the rug out from
    under them and start charging them is ludicrous. If this were to
    happen, I can honestly see a major fork in the GPL happening.
    • by Anonymous Coward
      don't be an idiot, who would they pay? the article is about releasing modifications. frankly, the article smells fishy, but whatever.
    • by gclef (96311) on Friday April 08, 2005 @06:05PM (#12181893)
      I'll go farther: This is an incredibly stupid idea. It will only convince businesses that the Linux/free software folks really are the communist, business-hating zealots that they're painted as.

      If you're not distributing your work, there's no reason why you should be forced to open your code or pay some silly fee.

      I hope this whole story is a troll...I really do.
    • by pmike_bauer (763028) on Friday April 08, 2005 @06:15PM (#12182017)
      "The idea that you can get someone hooked on software, and then pull the rug out from under them and start charging them is ludicrous."

      These implications of the proposed GPL3 are certainly troubling. How is this different from "evil commercial vendor lock-in"?

      I'm not trolling...just hoping that this interpretation of GPL 3 is wrong.

      • That interpretation is wrong. If something is licensed under GPL v2 now, you cannot take that away, unless a court rules that GPL v2 is invalid (which was the original justification for including the "or later" clause). So noone is pulling the rug out from anyone.
    • by bug1 (96678)
      The buisness world is on a very good wicket. They are getting us to make the tools they use to enslave us.

      Just becasue something is free as in beer doesnt mean it is devoid of all responsability from the users behalf.

      If a company's buisness model is dependent on free software then its in their companies best itnerest to be very generous to the programers who maintain the software they require.

      It sounds like the GPL v3.0 is trying to make the commercial world a little bit more responsable.
    • The idea that you can get someone hooked on software, and then pull the rug out from under them and start charging them is ludicrous.

      That's not what's being proposed at all. Not one line of code you currently possess that is under the GPL will be subject to v3 of the GPL unless you deliberately choose to do so.

      There's a case to be made if the GPL strays from it's free nature, but that's true of every license, not just Free/Open Source licenses. Where the GPL (and others) stands apart from proprietary lic
  • FSF? (Score:3, Funny)

    by panxerox (575545) * on Friday April 08, 2005 @06:00PM (#12181825)
    Shouldent that be FAIBFSF? Free as in beer free software foundation?
  • Bad, bad, bad! (Score:4, Insightful)

    by TinyManCan (580322) on Friday April 08, 2005 @06:00PM (#12181828) Homepage
    I _REALLY_ hope this idea falls flat on its face. These companies are the brightest hope for Free Software. Screwing them over is incredibly idiotic.
  • by n1ywb (555767) on Friday April 08, 2005 @06:00PM (#12181831) Homepage Journal
    Pay money to whome? When? How?

    I think this proposal is crazy. If you use software as a part of running your business, that software is benefiting you and indirectly providing services to your customers, even if they never see it. So where do you draw the line?

    • by winkydink (650484) * <sv.dude@gmail.com> on Friday April 08, 2005 @06:04PM (#12181889) Homepage Journal
      Why you'd pay it to the FSF, of course. They'd administer the money, funding projects as they see fit. Kind of a Politburo for the Software Community.

      BTW, when does Stallman's Macarthur Foundation Grant expire?

      For the humor impaired: It's a joke son.
    • by PHPgawd (744675) on Friday April 08, 2005 @06:11PM (#12181965)
      The payment model could be based on a simple royalty each time each subroutine is accessed. The easiest way to implement this would be an opaque wrapper on all GLPed APIs, which in turn could be easily implmented in a new version of gcc.

      The wrapper code could then count the number of calls, the function name called, and the company calling them, and this in turn could append a text file that is stored somewhere and automatically sent once a month to the FSF which will then use the file to send the company a bill. The FSF would then distrubute the money to each of the authors of the given API calls.

      I will volunteer to make the necessary modifications to gcc. This of course will require all GPL'd code to call my APIs, which will of course be the first to implement this new system.

  • by qewl (671495) on Friday April 08, 2005 @06:00PM (#12181834)
    But who gets paid the money and who determines how much? Can rates increase as GPL'd software develops? Is money paid to Richard Stahlman [sic] to distribute through the FSF? Seems an easy way to get around this anyway would just be to have small closed-source scripts that would only be called from the GPL'd code. After all, what's wrong with that? (so long as they're not redistributing their code, just letting use of it as a service)
  • Im speechless..... (Score:5, Insightful)

    by Richard_at_work (517087) <richardprice AT gmail DOT com> on Friday April 08, 2005 @06:00PM (#12181837)

    "If you look at the market, Yahoo, eBay, IBM, Amazon, Google have all sunk millions into the GPL infrastructure," Olson said. "Not only are we changing the rules, we are changing them retroactively. With the new way, it lets the customer pay with either their source code or with their wallet."

    Basically, in any other language: 'Now youve had time to build a good infrastructure on the current rules, prepare to be shafted'. If this comes to pass, then in my mind they are no better than Microsoft changing EULA terms with a service pack. Now that there are major companies with an infrastructure built on GPL software, the FSF are looking to essentially move the goalposts and if this is applied retroactively to current code (which from the articles wording I think it will be) then I personally think that its going to do more harm to the GPL community than benefit it in code donations as companies scramble to move away.

    Please someone tell me that they cant do this retroactively, that its impossible under the current GPLv2 terms.

    • Please someone tell me that they cant do this retroactively, that its impossible under the current GPLv2 terms.

      You can't, at least in the U.S. If you specify certain terms of copyright you have no right in the future to restrict those terms further - period. You can impose stricter copyright terms on new products, or new versions of products, but not on the products released under the older copyright.

      Once a copyright is specified, you can only loosen the restrictions, not tighten them. Unless, of cour
    • by kbmccarty (575443) <kmccarty@g[ ]l.com ['mai' in gap]> on Friday April 08, 2005 @06:12PM (#12181980) Homepage Journal

      Please someone tell me that they cant do this retroactively, that its impossible under the current GPLv2 terms.

      Your wish is my command. If you look at source code to any GPL-licensed program, you will see something like:

      // This program is free software; you can redistribute it and/or
      // modify it under the terms of the GNU General Public License
      // as published by the Free Software Foundation; either version 2
      // of the License, or (at your option) any later version.

      Hence any code that was licensed under the GPL before version 3 is released may still be used under the terms of the GPL version 2, at the option of the recipient, not the author. Actually, a number of current projects, including Linux, are licensed GPL-2 only and may be impossible ever to convert to a higher version (permission would be required from too many people to reasonably track down). Hope that helps.

      • future submitted patches can be GPL v3 only. the original author already gave permission for all of their work to continue and be relicensed under v3 in the future if the overall project decides to go that way. if that author doesn't like it they don't have to contribute to the project that has moved on to operate under the new version.
    • They can't do it retroactively.

      The current licenses say, and I quote:

      "you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version."

      Even assuming every open-source project immediately changes that to "version 3 or later", the source that was available *before* the change is still licensed under version 2. The company won't be able to upgrade to later v3-licens
  • That if you are interested in seeing the software spread and be used, that you wouldnt do something like this. Kinda like biting the hand that feeds you.

    I know that if I ran a company and I was compliant with a license, and the license changed to where I was penalized, I'd be less interested in using that software, but thats me.

    (no, I didnt read the article)
  • by FreeLinux (555387) on Friday April 08, 2005 @06:01PM (#12181850)
    Just the implication of this terribly and inaccurately worded headline will be devastating to open source. This article is intentional flamebait. Zonk, please try to be a bit more responsible with the articles you post.

    Regardless of what GPL 3.0 turns out to be, developers are not forced to use it. They can continue to use GPL 2 if they wish, just as they can choose to use a BSD license, Apache license, creative commons, or any other license of their choosing. Furthermore, software that has already been released under GPL 2 cannot be retracted, it remains available under GPL 2 forever.
  • by Anonymous Coward
    Such a clause will really ensure BSD never dies.
    • by Anonymous Coward
      Well, the BSD license at least.. it will be the last little reminder.
  • Ingenous (Score:2, Interesting)

    by onyxruby (118189)
    Ingenous, charge people who don't redistribute free software! I can't think of anything Microsoft has considered better news in a long, long time. Free software is no longer free. It's sorta free, well could be free, under the right circumstances is free. Ah fuck it, let me get a lawyer....
  • by vyrus128 (747164) <gwillen@nerdnet.org> on Friday April 08, 2005 @06:03PM (#12181873) Homepage
    The submitter isn't clear about the fact that this would not apply to everyone who changes software for commercial use but does not redistribute. This applies ONLY when "redistribution" of the software sort-of-occurs, because the software is used to provide a service. For example: any open source software Google uses in its search engine interface does not count as "redistributed", even though it _interacts_ with users of Google, because no actualy _binaries_ were shared with those users. For another example, if I modified the GIMP and then let people use my modified version over the Internet, but not on their machines, I would not have "redistributed" my modifications. This is considered by many to be a "loophole" to be closed.
    • by ajs (35943) <ajs AT ajs DOT com> on Friday April 08, 2005 @06:22PM (#12182078) Homepage Journal
      Right, so for example, a friend of mine works for a financial firm which he reports makes use of (and even contributes to) OSS projects. That firm provides a service based on these products (both electronic and off-line transactions that they perform as part of their core business). If they attempt to make this retroactive, I assure you that the world will come crashing down on the FSF. Thousands of firms around the country will sue them OVER NIGHT, and honestly, I'd be more likely to donate to the defense of those firms than the FSF (regardless of the fact that I'm a huge fan of the GPLv2).

      This is deeply irresponsible. Any project that ships software under the GPL is going to be spinning their wheels for months over this, and the Microsofts of the world just got a huge weapon to use against OSS usage. After all, now they can say that using GPL software not only costs you in terms of the usual TCO metrics, but there's a potentially hidden and as-yet-unknown cost that can be applied to retroactively!

      Grrr!
      • by Elwood P Dowd (16933) <judgmentalist@gmail.com> on Friday April 08, 2005 @07:07PM (#12182460) Journal
        If they attempt to make this retroactive, I assure you that the world will come crashing down on the FSF.

        Ditto if they attempt to murder strangers in public. Fortunately, neither will happen. If you are using a product under the GPL 2.0 license, then it is yours to use under the GPL 2.0 license.

        If someone releases updates to a GPL 2.0 licensed product, but those updates are only available under the GPL 3.0, and you do not want to abide by the restrictions in the GPL 3.0, then you might be SOL for those updates. That's about the worst thing that can happen.

        So chill out.
  • by karmaflux (148909) on Friday April 08, 2005 @06:03PM (#12181875)
    Why is everyone always saying "Software should be free, unless you're a business, in which case, get your checkbook"? The GPL shouldn't mandate anything except that code be made available. I think corporations should be entitled to the same rights and privileges as private citizens.
  • ... that the BSD-license paranoids might have been right?

    Ooops.

    Bad idea, Eben. It might make a good optional license (the Greater GPL?) but that's not something for the core GPL.
  • Isn't this 7 days late?

    I mean, sure, they don't distribute all the work they do, but what they do is a trade secret, and not copywrited.

    I think they need to strengthen the GPL more before adding more to it. I mean cisco and lots of other vendors sells products running GPL software (like linux) and there is no way to get the source code.
  • Call it FUD (Score:5, Insightful)

    by thephotoman (791574) on Friday April 08, 2005 @06:06PM (#12181903) Journal
    Frankly, I don't think this is even the case. Yes, the FSF has been talking about making a GPL 3.0 for a while now, but the proposal they're offering makes no sense.

    The problem is that it infringes on one of the four freedoms, specifically the freedom to use. If such a provision were to find its way into any license that made it so that companies and individuals were not subject to the same terms, the license would be both discriminatory and non-free.

    Simply put, this is somebody making FUD about the GPL. Don't buy into it for a second.
  • looking at a proposal where companies would be required to pay money if they use GPLed software,

    Wow, talk about sensationalism. Or maybe I can't read. But I did RTFA, and no where did I see mention of anybody having to pay anything for using GPLed software.

    Instead, what I did find was an article which seems to imply that the FSF is trying to further alienate themselves from the real world by effectively prohibiting GPLed software from being used by for-profit organizations, irregardless of whether or
    • This quote from the article:

      "If you look at the market, Yahoo, eBay, IBM, Amazon, Google have all sunk millions into the GPL infrastructure," Olson said. "Not only are we changing the rules, we are changing them retroactively. With the new way, it lets the customer pay with either their source code or with their wallet."

  • by Cylix (55374) * on Friday April 08, 2005 @06:08PM (#12181936) Homepage Journal
    I heard the next version of the GPL will require everyone to throw pies at Bill Gates.

    At least, that's what I heard anyway.

    When whatever changes come up, they will be reviewed and we can rant and rave about it at that time.

    Nothing is founded, no concrete written agendas were tossed out by the article and all we have is a hypothetical situation that would be very different from the current model.

    Like many have said and will say, it's usually a good idea to specify the version of the GPL in which you release software. Unless you really don't care what kind of changes are made at a later date.

    I'm not saying all of this isn't going to come true, but at this juncture we could very well be required to throw pies to comply with the next redistribution agreement.
  • by Andy Dodd (701)
    Do we want to trust a source that can't even bother to check their articles for typos?

    Who is Richard Stahlman???
  • Don't get Dramatic (Score:4, Insightful)

    by MattW (97290) <matt@ender.com> on Friday April 08, 2005 @06:10PM (#12181955) Homepage
    Moglen has stated: "I do not believe that we will be reach consensus on this front, so I believe the license will have to accommodate options as to the question of Web services, but this must be squared with the ideological pursuit of freedom". [ref. [eweek.com]]

    And frankly, it's not really a loophole. Web services are not distributing software, they're running a service using software. That's obviously open to interpretation, but I haven't ever heard anyone distribute under the GPL and complain about someone using their software as a web service. There has been at least one derivative license which has addressed this issue.

    In the end, GPL 3.0 will likely provide an optional provision which will 'trigger' GPL source distribution requirements for a web service, at the option of the copyright holder; that is really the best choice. Rather than getting into an enormous philosophical debate over whether the idea is "good" or "bad" or "punitive" or whatever, let's simply have two clear licenses and give the option to the copyright holders to decide under what terms they will license their property.
  • My god, I hate it when HP,MS or whomever does this. Now the GPL goes this way, UGH. It's not the fact that they are changing the GPL, that's bad enough but the fact they are retroactivly changing is what makes it so bad. This is the kind of shit the the bad boys do. I hope this idea goes no where or you can see any the profesional devs go elsewhere really quickly.
    • Re:Pot meet kettle (Score:4, Interesting)

      by kbmccarty (575443) <kmccarty@g[ ]l.com ['mai' in gap]> on Friday April 08, 2005 @06:40PM (#12182246) Homepage Journal

      My god, I hate it when HP,MS or whomever does this. Now the GPL goes this way, UGH. It's not the fact that they are changing the GPL, that's bad enough but the fact they are retroactivly changing is what makes it so bad. This is the kind of shit the the bad boys do.

      "They" can't change the license retroactively. Any software licensed under GPL is either "version 2 only" or "version 2 or later, at your option". So, any GPL software that exists at the moment the GPL-3 is unveiled, you may continue to use as long as you want under GPL-2 terms. Of course, this will not be true of new or updated software released afterward under "version 3 only" or "version 3 or later" terms.

      I hope this idea goes no where or you can see any the profesional devs go elsewhere really quickly.

      If the GPL-3 turns out to be as implied in the article, I completely agree with you. And I think that most free software authors will as well. If the terms of GPL-3 are that bad, no one will use it and it will quickly become irrelevant - nothing to worry about.

      But let's wait until we actually see a draft of a proposed GPL-3, not get upset over third-hand rumors.

  • ka-ching ! ! !
    • Airbus uses Berkeley DB for an application to display aircraft technical data.
    • Akamai uses Berkeley DB in its global content delivery network.
    • Alcatel, one of the world's leading telecommunications companies, uses Berkeley DB in its router and network management services products.
    • Amazon.com uses Berkeley DB as a fast cache for several critical parts of its customer-facing e-commerce website.
    • A9.com, a subsidiary of Amazon.com, uses Berkeley DB as the foundation for the HistoryServer in
  • by Doc Ruby (173196) on Friday April 08, 2005 @06:12PM (#12181982) Homepage Journal
    That new constraint would be a totally different principle for the license. The current principle is that if someone gets an executable from you, they also can get the source code, just as you got the source code from which you made your executable. Just using the source code, or customizing it for your own use, doesn't require distributing the source. The new principle would be requiring anyone who customized the source to release all customizations.

    This will stop many programmers, many of the best programmers, from using source code under that license. We could no longer keep any of the value of the software we created to ourselves. What is a "web service"? Is my email-processing CGI a "web service"? Any software in the same workflow as any other software under this license would have to be released. So many developers won't make small customizations, because that would force us into the source code distribution business, with all its overhead. Or we might just ignore that provision, or the whole license, en masse.

    The GPL is successful because it is a fair contract, even though it's revolutionary. Its enforcement teeth are rarely tested, because it's so close to an equitable agreement among peers. Which has resulted in lots of value contributed by profit-driven organizations, despite the claims of many that the license is anticapitalist. Upping the ante, to require private customizations to be published, could stop the rising tide of acceptance that is pushing GPL to be the default, and any proprietary license to be radical. And then the caution it would inspire: investing in GPL'd software might force acceptance of ever-more demanding licenses, like a GPL4.0 that required redistribution of even software that wasn't changed at all, just to get users "to pull their weight".

    The GPL2.0 isn't broken. Let's not "fix it" in a way that could destroy its success, and our chances to benefit from one another's work without onerous burdens.
  • they're looking at a proposal...

    They are just looking at it; just like they might be, for example, looking at a proposal to charge $699/CPU (dammit, why does that sound familiar?).

    If Google (for example) distributes GPLed software in their Appliance, then that is, by definition, redistribution and hence they must release the source to the customer of the said box.

    On the other hand, if Google uses GPLed software to provide a service, then I see no need for them to go about distributing the source too.

  • The entire basis of copyright is that information should not be free and that the creator/owner of information should have the right to charge what they want for copies of their creation. In contrast, the OSS movement argues against this on the basis of moral error (information should be free/not owned), impracticality (technology means that information wants to be free despite owner's wishes), or alternative business model (information is offered for free, but service is not).

    I can see that Google and
  • TFA is a little skimpy, i guess since they are still working the issue. I don't fully understand what it is trying to say.

    What does 'distribute software as a service' mean in this context?

    Is it that the GPL will say that the HTML/javascript web pages that Google serves up are derived works of the underlying (modified by Google) GPL'ed software? The underlying GPL'ed server binaries aren't sent out to the client. In effect, do they want to say some types of output of GPL software is a derived work?

    Or is
  • I have a small but nice piece of GPL code out there that I do not want to get paid for, neither do I want it to be used by people that give their clients less freedom than I did when I decided to distribute it.

    I mean, it's out there for anyone to use gratis, however it does not mean that I am allowing anybody to pimp it, that is why GPL2.
  • I want to see someone defend both this empire-building agressive new GPL license burden, and deny the claims that Linux extremists threaten their own success.
  • by Linuxathome (242573) on Friday April 08, 2005 @06:32PM (#12182185) Homepage Journal
    With the advent of the internet, it's now easier for computer programmers to freelance (see the well known Rent-a-Coder website). My impression is that under the current GPL, a programmer possibly could legally do contract work by adding to a GPL'd project [linuxathome.com] without being required to release the work -- so long as the buyer will never distribute the program via CD, floppy, or internet (i.e. keep the program "in-house"). Someone, correct me if this is not true.

    It appears that closing this loophole will also close the doors for programmers to freelance in this manner. That is they won't be able to sell their programming service of enhancing a current GPL'd project -- unless, of course, the solicitor agrees to either pay or release the code. The other option is to force the programmer to pay the GPL fee and roll this expense into the contract costs. I think this issue has to be debated and discussed at length, because we can't go about and make a gut decision of saying this provision is a good addition to the GPL just because we want to make big companies like Google, Amazon, etc. pay. After all, it could affect freelance programmers -- this could very well be their bread-n-butter. Let's not get into the question of whether personal enhancements of GPL'd programming tools (e.g. IDE's) are required to be paid or released under the proposed provision, that could be a sticky situation.
  • by renehollan (138013) <rhollan@ c l e a r w i re.net> on Friday April 08, 2005 @07:00PM (#12182403) Homepage Journal
    The problem with servers and services is that it muddies the line between program linkage and not.

    Currently the GPL forbids redistribution of GPL-covered code linked with other code to form a larger work, when that other code is not also distributed under the GPL: the whole larger work must be so distributed. It is a "derived work", and copyright law is clear on this: one can not redistribute derivatives of copyright works without permission. It's that permission that the GPL grants, when you abide by its terms.

    Of course, you are not prevented from redistributing GPL code *aggregated* with non-GPL code.

    The problem arises from distinguishing between derived and aggregated works: what if I distribute a GPL app that I write, with a bunch of GPL shared libraries that I didn't write (complete with source), and a nonh-GPL proprietary library, essential for the application, that I did write, without which the application is useless.

    Is that an aggregate, or a derived work of all the GPL code I did not write?

    On the one hand, there's no (legal) requirement that my app even run, and, since I distribute source, one can implement a version of the non-GPL library. The fact that it works when my non-GPL library is installed in the right directory (LD_LIBRARY_PATH, anyone?) is a happy coincidence. Redistributing my GPL app, with modifications is fine, but you can't redistribute my "essential" non-GPL library, rendering the app rather useless without it. You add value to the app, I license more copies of the essential library.

    On the other hand, the proprietary library, the foreign GPL libraries, and my app together, constitute a derived work of the GPL libraries, and must be redistributed en masse under the GPL.

    If the libraries are statically linked into a monolithic executable, there is little argument that the whole source must be redistributed. The situation is more controversial if a mere aggregate of files is distributed: some argue that not distributing the proprietary library under the GPL is a violation, other's don't.

    I tend to believe that if the GPL-covered peices can be redistributed indivudually, without the proprietary library, regardless of whether the result is useless, there is no violation: the fact that the code is actually only useful when the proprietary library is present is but a happy coincidence. What if I distributed a GPL version of the library, and offered a proprietary replacement, with far better performance?

    This controvery gets even muddier when one considers alternate ways of effecting program linkage. In effect, the "functional derived work" exists only at run-time, and, indeed, the maner in which the parts of the aggregate are combined, can, itself, be the subject of restrive licencing, and patent. What about linkage via a remote procedure call mechanism? Y'all remember RPC/XDR over TCP/IP, right? Suddenly, self-assembling functional derived works become a reality. Protocols like SOAP, used to support "Web Services" exacerbate the problem.

    This leaves a big, gaping hole in the GPL: socket wrappers were a common "hack" to "get around" the GPL: just layer a RPC mechansism around the proprietary library, and a GPL wrapper to call it from the app, and you were all set. And, that didn't even address the issue of inter-machine communication: 127.0.0.1 and Unix sockets are ubiquitous.

    Of course, the minute one's app "links" with proprietary code on a different server in this manner, the GPL loses all force, for one did non distribute the proprietary part.

    It is this area that the GPL v. 3.0 tries to address, IMHO, perhaps by more differentiating between linkage and aggregation. This can be done, of course, but then the license starts to lose some of its roots in copyright law.

    The problem with such an approach, though, is, what constitutes linkage: does an exchange of HTTP requests and responses? What if there is ultimately GPL code satisfying that request? Surely, if I use a program provided by a vendor

  • by nuggz (69912) on Friday April 08, 2005 @07:33PM (#12182688) Homepage
    It specifically states there is no restriction on running the program, or what you can do with the output.
    Changing from a distribution license to a usage license is a VERY significant change in the spirit of the license.

    Quote Last para, Term 0
    The act of running the Program is not restricted, and the output from the Program
    is covered only if its contents constitute a work based on the Program


    Term 9
    Such new versions will be similar in spirit to the present version,
  • by Indomitus (578) on Friday April 08, 2005 @07:43PM (#12182808) Homepage Journal
    Wow, thanks FSF folks. You think the minor and stupid threat of legal action against Linux using corporations by SCO chilled use of Linux? Even talking about a change like this could do far more damage. It is incredible that they would even talk about something like this. If I'm a company like Google, why would I go forward using GPL software when the people in charge of the license talk about changing it to make me pay when I'm not redistributing my code? What's next? I run a website on Apache, I have to pay the FSF for the privilege? Sheesh.
  • by gumtu (206776) on Friday April 08, 2005 @07:51PM (#12182892)
    Although the intent behind the GPL is laudable, and the wording of the GPL in its current form achieves this goal reasonably well, it is always a bad idea to have a small cabal of minimally accountable people control such an important lever.

    The GPL is a single point of control over the vast majority of the FOSS movement (~60-75% of all projects according to Wikipedia). The wording of the GPL impacts not only the fate of the FOSS movement itself, not only the fate of the work product of each individual participating in this movement, but also the fate of all the companies (large and small) that have chosen to assume the risk of depending on this software.

    Given the GPL's extreme importance to such a large and growing audience, we should all take a hard look at who really controls it. The GPL is controlled by the FSF. The FSF is a 501(c)(3) non-profit; it has a board of directors who have responsibility for oversight of the President who runs FSF day-to-day (RMS is the president). The board of directors is elected by "Members" (NOTE: If you join the FSF off their website, you are an Associate Member which is a NON-VOTING position). I'm not sure how one gets to be a full Member.

    Now let's not kid each other here: We all know how sketchy the oversight of a Board of Directors can be; we all know that groups of theoretical "equals" can be strongly influenced by a small number of strong personalities; and we all know that "Strong Personality" is a very accurate two word description of RMS (and I doubt Eben Moglen is far behind). I think those among us who are objective (and especially those among us who have personally interacted with RMS) can agree that RMS is on a personal jihad and he's using the force of law to achieve his ends. The question is, do we all want to participate in that same jihad?

    Of course, one can respond by saying: "There is a competitive market for licenses - no one is forcing developers to use the GPL." This might be true on paper but it's not true in practice. We're software developers, not lawyers. I'm willing to bet that the vast majority of developers release their work under the GPL simply because "that's the one I've heard the most about and that's what everybody else does", and NOT because they researched other available licenses and arrived at an understanding of their implications, and especially an understanding of who controls these licenses.

    So where does that leave us? We have a single pseudo-fanatic who has substantial control over the single most important component of our movement. This should scare the shit out of all of you.

    Me? I'm going to either change to a non-GPL license or stay with GPLv2.

    It's funny how the FOSS movement parallels Russian history: RMS (Lenin) started a revolution to overthrow proprietary software (the Czars). But the new regime became a dictatorship (Communism). It's now time for the proletarians (FOSS developers) to revolt again and achieve self-determination (free, accountable, and transparent democracy).
  • Olson should know. He is one of a select few looking to review the current GPL and recommend updates for the public review process, which he says should happen before the end of the year.
    Right, so Mike Olson is one of an infinite number of people who can read the current GPL [gnu.org] and recommend updates by mailing licensing@gnu.org [mailto] for public review. Obviously this makes him an insider. (Congratulations! If you're reading this, and can click or right arrow on two links, you're an insider too!)

    Perhaps he's just managed to read the Affero General Public License v1 [affero.org] and has decided that that's the way that the GPL v3 is going to look? But apparently he hasn't already read the coverage of this rather crappy license that debian-legal gave in 2003 [debian.org] and then informed the FSF (and RMS), explaining that it couldn't possibly be DFSG Free, [debian.org] let alone satisfy the 4 freedoms?

    Oh, right. Must not have actually checked all that out. Gee, does Mike Olson even use the GPL at all? Why would he be reviewing it anyway? Well, lets see: hrm... this sure looks like the 3 clause BSD license to me [sleepycat.com]. Yerp. No GPL in sight at all. Ok, so someone who doesn't even use the GPL, (to my knowledge) isn't a lawyer, and isn't a prominent member of the copyleft side of the Free Software movement is reviewing a license that no one else has seen?

    I mean, I can understand slashdot editors missing this bit of trivia in their rush to approve/reject a story... but surely Michael Singer at internetnews would have bothered to actually check if Mike Olson was the "insider" he was claiming himself to be?
  • by Edmund Blackadder (559735) on Saturday April 09, 2005 @12:26AM (#12184748)
    I am very surprised that Slashdot would actually approve such a misleading story. Misleading stories have appeared on Slashdot before, but at least when it comes to free software, the stories used to be truthful and informative. Well , this is an unfortunate exception.

    First of all the story purports to tell what Eben Moglen is doing but mostly includes quotes from that Olson fellow. The obnly Moglen quote basicly says "let me finish the draft first and then we can discuss it". Then the article continues on pure speculation of what Moglen might be thinking.

    Second, the slashdot blurb completely misrepresents the actual referenced article. The shashdot blurb makes it look like the main purpose of the FSF is to hit up google, yahoo, etc. for money. In reality the whole damn controversy is whether the code should be released or not.

    Of course if there ever emerges a legal requirement to release the code and the code is not released the FSF might be entitled to damages -- thats just the way the US judicial system works. But that does not mean that the FSF are looking to hit up these companies for money. In every dispute so far the FSF has made it quite clear that they will forgo money damages when the code is actually released.

    Now whether GPL derived code used for providing internet services should be released is an very interesting (and increasingly important) issue. However, this slashdot article completely misses the whole issue by making it look like the FSF is just trying to get rich on Google's behalf.

God doesn't play dice. -- Albert Einstein

Working...